Nonvalvular atrial fibrillation (NVAF) is an independent risk factor for ischemic stroke and leads to a fivefold increase in mortality compared with normal subjects [1]. NVAF is one of the important causes of stroke and cardiovascular events worldwide. According to statistics, the risk of stroke in atrial fibrillation (AF) patients is five times higher than that in the general population, and the incidence of NVAF and associated stroke risk are significantly increased, especially in elderly patients. Pathophysiological mechanisms of NVAF involve atrial electrophysiological changes like structural remodeling, fibrosis, and electrical conduction abnormalities, leading to ineffective atrial contraction and arrhythmia. Structural changes in the atrium, inflammatory responses, and autonomic nervous system effects also contribute to the development and maintenance of NVAF, with underlying diseases and chronic inflammation playing key roles in promoting the arrhythmia [2].

Due to ineffective and irregular rapid atrial contraction and diastolic dysfunction, blood circulation through the left atrium or left atrial appendage can lead to stagnation, resulting in the formation of left atrial thrombus (LAT) or left atrial appendage thrombus (LAAT) [3]. In the evaluation of biomarkers for left atrial appendage thrombosis, MPV (mean platelet volume), NT-proBNP (N-terminal pro-B type natriuretic peptide), RDW (red blood cell distribution width) and CHA2DS2-VASc score have shown certain potential. In-depth study of these biomarkers can help reveal their true value in clinical applications and inform future therapeutic strategies [4]. LAT or LAAT serves as the source of embolus in AF [5]. Therefore, thrombus resolution is crucial in preventing embolic events in AF patients with LAT or LAAT.

Anticoagulant drugs in the clinical application for AF can be categorized into two groups, namely, warfarin and vitamin K antagonists (VKAs) and non-vitamin K antagonist oral anticoagulants (NOACs) such as dabigatran, rivaroxaban, and apixaban. Before the introduction of NOACs, studies had demonstrated that warfarin reduced the risk of stroke by 66% [6]. However, patients with AF often struggle with compliance and coordination due to the influence of food and drug interactions, the need for the frequent monitoring of coagulation, and a narrow treatment time window associated with warfarin use. Recent studies have confirmed that NOACs are more effective than warfarin in preventing stroke and reducing bleeding events in patients with NVAF [7, 8, 9]. Consequently, NOACs are now recommended in the guidelines for the prevention of embolic events in patients with NVAF [10].

In AF patients with LAT or LAAT, the primary challenge is dissolving the thrombus. Karwowski et al. [11] have confirmed that a significant proportion (approximately 58%) of the population is affected by LAT/LAAT. Although case reports [12, 13] and clinical research [14, 15, 16] have shown that NOACs are as effective and safe as VKAs in treating AF patients with LAT/LAAT, there are limited data on the efficacy and safety of LAT/LAAT resolution with NOACs. The optimal regimen for thrombolytic therapy in AF patients with LAT/LAAT remains unclear. Therefore, this study aimed to systematically evaluate the efficacy and safety of NOACs and VKAs in treating LAT/LAAT through meta-analysis. The results of this study could provide valuable insights for the clinical management of AF patients with LAT/LAAT.

We searched both Chinese (CBM (China Biology Medicine disc), CNKI (China National Knowledge Infrastructure), VIP (Chinese Technology Periodical Database), and Wanfang) and English (PubMed, Embase, Cochrane Library, and Web of Science) databases without imposing any restrictions on the study type. The databases were searched from their inception to June 30, 2022. A manual reference search was conducted to meet inclusion criteria and enhance the reliability of the study conclusions. A search formula using keywords was structured as follows: (“new oral anticoagulant” OR “non-vitamin K oral anticoagulants” OR rivaroxaban OR dabigatran OR apixaban OR edoxaban) AND (“atrial fibrillation” OR “left atrium” OR “left atrial appendage” OR thrombus) AND (“resolution” OR “dissolution” OR “decomposition” OR “dissolve” OR “thrombolysis”) AND (“randomised controlled trial” OR “controlled clinical trial” OR randomised OR randomly OR trial). The search strategy used in this study is outlined in Table 1. The reporting of the study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).

We performed this meta-analysis of four RCTs and eight cohort studies with 982 patients to investigate the efficacy of NOACs vs. VKAs in the resolution of AF with LAT/LAAT. The results showed that the thrombolysis rate of NOACs was higher than that of VKAs; NOACs did not increase the risk of stroke or bleeding during the application and had more obvious advantages for thrombosis treatment of NVAF.

The incidence of LAAT in AF patients ranges from 0.5% to 14% [31, 32]. Antithrombotic therapy is crucial in patients with LAT, as it is an absolute contraindication to radiofrequency ablation. There are limited studies on thrombolytic treatment for NVAF with LAT/LAAT, and there have been no large RCTs to guide treatment. This study directly compared the efficacy and safety of NOACs and VKAs for thrombolytic therapy in NVAF with LAT/LAAT. Warfarin, a representative VKA, effectively reduces stroke risk but has limitations in clinical use, such as food and drug interactions, frequent monitoring of international normalized ratio (INR), and individual dose–response variability [8]. Current studies suggest maintaining an INR between 2.0 and 3.0 for effective anticoagulation without increasing bleeding risk [33, 34]. Most studies in the literature have used an INR range of 2.0–3.0 for the warfarin control group. However, the RCT by Huang Fangfang et al. [26] divided warfarin into low-intensity (INR 1.5–2.0) and standard-intensity (INR 2.0–3.0) groups, and they found that the effects in the low-intensity group were not superior, but there was a reduced risk of bleeding.

Dabigatran, a direct thrombin inhibitor, has demonstrated a lower prevalence of left heart thrombosis than warfarin in combination therapy [35]. Two common clinical doses of dabigatran (110 mg bis in die (BID) and 150 mg BID) have different efficacy and safety profiles. The 110 mg BID dose is comparable to warfarin in terms of efficacy and reduces bleeding risk, while the 150 mg BID dose increases stroke prevention ability but has comparable bleeding risk to warfarin. Study of Lip et al. [36] focused on the thrombolysis effect of dabigatran and included studies using both doses. However, there is limited literature on dose grouping, preventing further verification of different doses’ effects on efficacy and safety. Dabigatran has shown advantages over VKAs in stroke prevention and reducing bleeding risk [8]. Ongoing clinical trials, such as RE-LATED AF-AFNET [35], are studying thrombolysis efficacy of dabigatran compared with warfarin, with results awaited. For patients over 80 years old, the recommended dose is 110 mg twice daily. This article focuses on dabigatran’s thrombolytic effects; however, due to insufficient literature on dosing variations, the impact of different doses on efficacy and safety remains unverified.

Rivaroxaban is a factor Xa inhibitor that effectively inhibits thrombin formation. Compared with warfarin, rivaroxaban has high bioavailability and is not affected by food and drug interactions. The ROCKET-AF [37] study has extensively evaluated the efficacy and safety of rivaroxaban in preventing stroke events. Previous studies, including the X-TRA study [38], have demonstrated the thrombolytic effect of rivaroxaban, making it a viable option for NVAF with LAT. The mechanism of rivaroxaban’s thrombolytic effect involves inhibiting thrombin release from blood clots and increasing fibrinolytic activity [29]. This study includes abundant data on rivaroxaban for LAT in NVAF patients. Subgroup analysis confirmed that rivaroxaban performs better than VKAs in resolving thrombosis in AF patients. Current expert consensus recommends a daily dose of 20 mg taken with food; no adjustment is needed for those over 65 years old. However, for elderly patients with nonvalvular atrial fibrillation and Ccr (creatinine clearance rate) of 15–30 mL/min, a lower dose of 10 mg/d is advised.

Apixaban is a direct oral factor Xa inhibitor, similar to rivaroxaban, that regulates thrombin synthesis in anticoagulation for AF. It is superior to warfarin in preventing stroke, reducing bleeding events, and decreasing mortality in AF patients [39]. Apixaban is primarily metabolized by the liver and excreted through the digestive tract and kidneys, requiring caution in patients with liver injury. The standard dose is 5 mg twice daily, but for patients with renal insufficiency, a reduced dose of 2.5 mg twice daily is recommended based on the AVERROES and ARISTOTLE [40] trials, especially in elderly patients, those with low body weight, and those with high creatinine levels. These studies have shown similar efficacy and safety of apixaban compared to warfarin and aspirin in patients with normal renal function and even renal failure [41, 42, 43, 44]. However, limited data are available on the thrombolytic effect of apixaban and its comparison to VKAs, with small sample sizes and no statistically significant differences observed. Understanding the thrombolytic mechanism of apixaban highlights its advantage in shifting the coagulation/fibrinolytic balance toward fibrinolytic activity.

Edoxaban is a newly approved direct factor Xa inhibitor, similar to rivaroxaban and apixaban. One of the distinguishing features of edoxaban is that it does not interact with the cytochrome P-450 system. This makes edoxaban a viable option for patients with AF who require oral anticoagulation and have interactions with the cytochrome P-450 system. Regarding edoxaban dosing, Weitz et al. [45] conducted a phase II trial comparing various doses against warfarin; results showed that edoxaban at doses of 30/60 mg/d had similar efficacy and safety profiles. However, data regarding its efficacy in thrombolysis are limited. Only one study [21] directly compared edoxaban to warfarin in thrombolysis, but it did not provide further comparative data on the thrombolytic effect. More research is needed to investigate the comparative efficacy and safety of edoxaban and VKAs in thrombolysis.

NOACs have demonstrated significant advantages in treating left atrial thrombi, with differences in efficacy influenced by factors such as pharmacokinetics and drug targets. A meta-analysis study indicates that NOACs like rivaroxaban and dabigatran show promise in managing non-organized cardiac thrombi early on, potentially as standalone treatments without the need for additional anticoagulants [46]. Compared to VKAs, NOACs offer more predictable pharmacokinetics and pharmacodynamics, ensuring better stability in anticoagulation effects and reducing thrombosis or bleeding risks [47]. Moreover, NOACs exhibit limited drug interactions, providing convenience for patients without the risk of reduced efficacy or increased side effects due to interactions. With fixed dosages and reduced need for frequent laboratory monitoring, NOACs positively impact patient compliance and treatment outcomes [48]. The rapid absorption post-oral administration allows for quick onset of action, especially beneficial in acute situations for timely management of atrial thrombi [28]. The diverse efficacy among different NOACs can be explained by their pharmacokinetic variations, including differences in absorption, distribution, metabolism, and elimination, which directly influence their efficacy in anticoagulation. For instance, Rivaroxaban’s high bioavailability and moderate half-life enable effective coagulation control within a day [45]. NOACs acting on different coagulation factors, such as direct thrombin inhibitors (like Dabigatran) and direct Factor Xa inhibitors (like Rivaroxaban), have distinct mechanisms impacting their clinical outcomes [49]. Individual physiological differences like age, weight, and liver or kidney function can also influence NOAC efficacy, potentially leading to varied treatment responses among patients [50]. Furthermore, variations in recommended dosages and administration methods among NOACs can affect their clinical efficacy and safety.

This meta-analysis divided the included studies into subgroups based on the follow-up time and study type. The results showed that NOACs consistently demonstrated better thrombolysis effects than VKAs in different follow-up times and study types. Dabigatran and apixaban had higher absolute thrombus resolution rates, although significant differences were not observed, likely due to the small sample sizes. Comparisons between dabigatran and rivaroxaban showed similar thrombolytic rates [51]. However, larger clinical trials are needed for further confirmation. During the treatment process, some patients underwent a switch in thrombolytic regimens; this led to successful thrombus resolution. Increasing the dose of the original drug or switching to another NOAC regimen (such as between rivaroxaban and dabigatran) showed good thrombolytic effects [20, 23, 52]. However, careful monitoring for bleeding risk is crucial when increasing the dose of the original thrombolytic agent.

Patients with increased left ventricular end-diastolic inner diameter are at higher risk of developing left ventricular thrombosis due to abnormal ventricular wall movement and the presence of blood vortex flow in the ventricle. A previous cohort study has shown a low risk of LAT [30]. Factors such as an increase in LAD and decreased LVEF, along with persistent AF, are associated with a higher risk of thrombus formation in AF patients and can also affect thrombus resolution [18, 53]. However, current research on predicting LAT factors is limited. This meta-analysis compared the data from three studies [18, 20, 21] on thrombolysis effectiveness and ineffectiveness. The results indicated that patients with ineffective thrombolysis had a decrease in LVEDd and an increase in LAD, with statistically significant differences compared with those with effective thrombolysis. However, the analysis did not reflect the differences in AF subtype and LVEF between the effective and ineffective groups due to the limited number of included studies, small sample sizes, and potential bias. These findings suggest the need for further investigation into the impact of persistent AF, LVEDd, LAD, and LVEF on LAT. Improving these factors may potentially enhance thrombus resolution in patients with NVAF and LAT.

In this comprehensive meta-analysis evaluating the efficacy and safety of NOACs and VKA in thrombolysis for Atrial Fibrillation with LAT/LAAT, minimal heterogeneity and high homogeneity were observed. The heterogeneity of each outcome index remained stable, unaffected by subgroup analyses or iteratively excluding literature. Despite variations in subject demographics and comorbidities across studies, excluding individual studies showed no significant impact on the consistent and dependable results obtained. However, notable heterogeneity was identified in the comparison of LVEF index between effective and ineffective thrombolysis groups, with limited data available for subgroup analysis, posing challenges in assessing heterogeneity. In the Nelles et al. [21] study, the absence of Left Atrial Appendage closure and alterations in anticoagulant therapy due to unresolved atrial thrombus underscore the need for cautious interpretation of combined index analysis results.

All of the included studies in this analysis used the CHA2DS2-VASc scoring system to assess the stroke risk in patients with AF. The results indicated that the incidence of embolic events was 3.4% with NOACs compared with 7.7% with VKAs, although this difference was not statistically significant (p = 0.06). Similarly, the incidence of bleeding events during thrombolysis was 8.1% with NOACs and 9.9% with VKAs, and no statistically significant difference was observed (p = 0.77). Only two studies [21, 22] reported on mortality events, and the analysis showed no significant difference in mortality risk between NVAF patients with LAT and those treated with VKAs.

NOACs have a higher thrombolysis rate in patients with NVAF than VKAs. The use of NOACs does not significantly increase the risk of adverse events such as embolism, bleeding, and death compared with VKAs.

Data availability is not applicable to this article as no new data were created or analyzed in this study.

GLM, YJY and JYL designed the research study of meta analysis. JW and YYY conducted the literature search and data extraction. GLM, YQL and TMG performed the statistical analysis and interpreted the results. All authors contributed to the drafting and revision of the manuscript, and approved the final version for publication. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Not applicable.

This study was supported by the National natural science foundation of China (Grant No. 82160077), the Self-Funded Scientific Research Project of Guangxi Health Department (Grant No. Z20211177), the General Program of Natural Science Foundation of Guangxi Province of China (Grant No. 2017GXNSFAA198129) and the Key Project of Scientific Research and Technology Development of Qingxiu District of Nanning, Guangxi government (Grant No. 2017027).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/RCM26055.